Third Draft, July 3, 2012

What Should Social Investors Invest in, and WithWhom?

by Richard SteinbergProfessor of Economics, Philanthropic Studies, and Public Affairs

IUPUI

ABSTRACT

I argue that social investors should invest in activities that cannot be reliablysustained through ordinary market transactions. These activities includecollective consumption goods, goods with nonverifiable quality, merit goods, andprivate goods that target groups cannot but ought to be able to, afford. Then Iconsider government’s role in social investment. Most of the paper is devoted tochoice of investment targets across institutional forms – for-profit, nonprofit, andhybrid organizations. Nonprofits and for-profits have distinct advantages anddisadvantages, and either might be the best home for specific kinds of socialinvestment. In contrast, I am skeptical of the claimed benefits of hybridorganizations and structures. I argue that LLCs (Limited Liability Companies),L3Cs (Low-profit Limited Liability Companies), CICs (Community InvestmentCorporations), B corporations, Benefit corporations, and Social Impact Bonds losethe efficiency advantage of traditional for-profits and do not enhance access tocapital over the nonprofit form. In addition, hybrids suffer from contract failureand some of the hybrid forms will have difficulty sustaining their social mission. Social investors should avoid existing forms of hybrid organization.

ACKNOWLEDGMENTS

I’d like to thank Kathleen Duffy Ballard, Wolfgang Bielefeld, John DeWolf, LijunHe, Robert A. Katz, Leslie Lenkowsky, Alex Nicholls, Rob C. Paton, Dennis R.Young, and participants in presentations at IUPUI and ARNOVA’s annualconference in Alexandria for helpful comments and suggestions. Research wassupported by a grant from the Center on Philanthropy at Indiana University.

I’d love to hear your comments, criticisms, suggestions, and perhaps favorite recipes. Best reached via email: rsteinbe@iupui.edu

Introduction

Some social investment supports neglected market niches that, once entered, are perfectly

sustainable through sales revenues, royalties, and rental payments. Social investors help these

ventures to germinate, but ongoing social investment is unnecessary because financial investors

take over. In this paper, I concentrate on social ventures that cannot be reliably sustained through

ordinary market transactions. I define the investment target, social enterprise, similarly to

Horton (2008), who wrote:

“[S]ocial enterprise provides, or tries to provide, things that society values

but which for one reason or another are not usually provided by private enterprise.

In most cases, this is because the social value of an activity is not easily captured

by the price or financial return for the service or good offered.”

This definition is well-suited to the “what” and “to whom” questions that structure my paper

because it focuses on difficulties in capturing value, rather than the nature of the value that is

created. The alternative, “organizations with social purposes,” has different meanings to

different people. Providing jobs and producing products that consumers want to buy are social

purposes, but by that definition every successful firm is a social enterprise. Narrower definitions

of “social purposes” leave many grey areas – is the discovery, production, and marketing of

various forms of insulin by Eli Lilly and Company a social purpose central to Lilly’s operation?

My definition excludes such cases. Lilly’s patents allowed it to capture the value of

insulin research, and drug sales capture the value of insulin production. Instead, I look at private

efforts to correct market failures resulting from collectively-consumable goods and services and

non-verifiable attributes of goods or services. I also look at provision of “merit goods,” goods

1

that, in the judgment of some members of society, are undervalued by other members. This

category also includes social marketing efforts that attempt to change habits and behaviors for the

subject’s own good. Finally, I look at affordability, where essential goods and services cost more

than consumers can pay. Social enterprises may subsidize these services or provide general cash

assistance. In sum, I argue that social investors should invest in three areas: market failures,

merit goods, and redistributive goods.

I briefly summarize government’s role in social investment, then consider the

organizational question: Should private social investments be directed to for-profit or nonprofit

organizations? I consider these forms as pure organizational types defined by the rights of

organizational owners. Full owners have three property rights: the right to control the use of

assets, the right to receive financial surpluses generated through use of assets, and the right to

transfer the first two rights at a market-determined price (Ben-Ner and Jones, 1995). Investors

own for-profits, possessing all three property rights. The for-profit form has subtypes defined by

the type of owners: worker-owned firms, producer and consumer cooperatives, mutual insurance

companies, and the like. In this paper, however, I mostly restrict attention to owners that care

about financial distributions rather than workplace conditions, services to member organizations,

or the characteristics of output. My arguments do not require me to distinguish corporations

from partnerships, proprietorships, or trusts but it is important to distinguish single-owner

organizations from multi-owner organizations. I label idealized single-owner organizations as

proprietorships whether or not liability is limited.

Nonprofits are owned by an appointed, elected, or self-perpetuating board of governors

2

that has the right to control the use of assets but not the other two property rights. Nonprofit1

corporations cannot distribute their surpluses to investors (donors) or those controlling the use of

the organization’s assets (board members) (Hansmann, 1980). Instead, nonprofit investors

accept “dividends-in-kind” (Wedig, 1994), deriving value from the organization’s activities and

outputs or outcomes. In contrast to for-profits, nonprofits that fail to generate market rates of

return will not be subject to shareholder pressures, derivative suits, or hostile takeover bids.

Thus, organizational missions are protected from the market for control.

The next section of this paper briefly explores the role of governments as social investors

and targets of private social investment. Then I consider private organizations as vehicles for

social investment, with sections on nonprofit and for-profit organizations. More recently,

societies around the world have experimented with nonprofit/for-profit hybrid forms, such as

Low-Profit Limited Liability Companies (L3Cs), Community Investment Corporations (CICs),

and “B Corporations” in an effort to combine the advantages of each pure organizational form. I

discuss whether these efforts are likely to succeed in the penultimate section, then conclude the

paper.

Government as Social Investor and Recipient of Private Social Investments

Although a comprehensive evaluation of the merits and limitations of government as

maker and recipients of social investments is beyond the scope of this paper, government’s role

I omit organizations with alternative sets of property rights (such as the first two but not1

the third) including some mutuals and cooperatives. I leave to future researchers the question ofwhether any of these alternatives dominate the organizational forms considered here.

3

is so huge that something needs to be said, if for no other reason than to place the merits of

private alternatives in context. Until recently, public economists thought of the state as the

remedy for all the shortcomings of the market. If markets under-provide collective goods,

government steps in, using taxes to finance direct provision or paying private organizations to

provide them. Governments regulate private organizations in ways that promote collective goods

like clean air and water. Governments deliver and subsidize merit goods. Redistribution is the

quintessentially government role, and social marketing is promoted through the public schools

and many other ways.

However, there are limits to the extent and nature of social investments by governments,

leaving a large role for private social investors. Governments are constrained by political forces,

whether they need to secure the approval of electorates, civil society elites, or the loyalties of the

military. This constrains their fiscal capacity for social investments and limits the kind of

investments that can be made. Securing government investments that benefit a minority of

citizens but are paid for by all taxpayers is difficult or impossible.

But even if government faced no political constraints, there is still virtue in facilitating

private social investments. Private investments provide pluralistic alternatives that help to confer

legitimacy to the inherently coercive activities of government. In addition, these investments

foster the voluntary impulse, promote innovation, and enhance the expression of individual

values in society.

Other social investors do make investments in government and government

organizations. For example, Andrew Carnegie funded the creation of 2,509 public libraries in the

United States and elsewhere (Burlingame, 2004). Alumni commonly contribute to the

4

endowments of public universities. Although it seems peculiar that someone would contribute to

entities that have the power to tax, Li et al. (2011) demonstrate substantial willingness to do so in

an experimental setting, and Irvin and Carr (2005) find evidence that such giving is concentrated

on particular types of issues at the local government level.

The Case for Social Investment in Nonprofit Organizations

Social investors provide capital to nonprofit organizations through their donations, grants,

loans, and contracts. Volunteer labor is, in effect, donated and sometimes labor is partially2

donated to the nonprofit in that workers are willing to accept a lower-than-market wage (Preston,

1989). Consumers sometimes overpay for nonprofit products (like Girl Scout cookies or posters

from the Metropolitan Museum of Art), in effect mixing a donation with a purchase. Members

pay dues that often exceed the costs of goods and services received, again, mixing a donation

with membership.

The extent of social investment in nonprofits ($303 billion donated to U.S. nonprofits in

2009 (Giving USA Foundation, 2010), plus the value of volunteering) is easy to understand.

Nonprofit organizations provide goods and services that address market failures or are

undervalued by consumers. Nonprofits offer financial aid, sliding scale fees, and charity care at

lower-than-market prices or for free. These socially-beneficial activities are embedded in

nonprofit mission statements and organizational cultures. Social missions can endure because

there is no market for the control of nonprofits. Nonetheless, nonprofits have, or are thought to

Although government purchase-of-service contracts with nonprofits typically provide no2

financial surplus, they sometimes cover capital expenses associated with the contract.

5

have, several shortcomings that should concern social investors. In this section, I first explain

the benefits then the shortcomings of the nonprofit form. A much more extensive discussion of

these virtues and deficiencies is provided by Steinberg (2006), although the discussion there is

not directed towards social investment.

Collective Goods.

Long ago, Weisbrod (1975) observed that many nonprofit organizations provide

collective goods, goods which, once provided, can be enjoyed simultaneously by many3

individuals. There are two kinds of collective goods – excludable and nonexcludable –

distinguished by the possibilities to restrict consumption to paying individuals. National defense

is nonexcludable – those who evade taxation are just as secure as those who pay taxes. An opera

performance is excludable – the audience collectively consumes the opera, but it is easy to charge

admission and thereby exclude some consumers. For-profits provide essentially no

nonexcludable collective goods because there is no meaningful way to charge for them.

Consumer payments for nonexcludable collective goods are purely voluntary, more like a

donation than a purchase. And as we shall discuss below, donations to for-profits are a rare

thing. In contrast, for-profits do provide excludable collective goods (like rock concerts), but

they choose prices that exclude too many consumers, from a pure “social efficiency” standpoint

as well as from an affordability point of view.

The literature often uses the name “public goods” instead of collective goods. In the3

history of economic thought, it was understood that governments would have to providecollective goods because markets would not, so the “public” moniker seems natural. But nowthat we realize that private nonprofit organizations often provide collective goods, and thatgovernments also provide non-collective goods, the older name is slowly disappearing.

6

Collective goods are provided by charities that help the needy, often in combination with

noncollective (rival) goods. For example, the food provided in soup kitchens is a rival good,

consumed by one person to the exclusion of others. But food provision collectively benefits all

those who care about hungry people (either out of sympathy, empathy, or a concern with property

value) in a nonrival way. Specific-disease philanthropies like the American Diabetes

Association, create collective knowledge that may provide a cure to specific sufferers (non-

collective). Nonprofit schools increase their students’ lifetime earnings (non-collective), but also

enhance later civic participation by their graduates (collective). Arts organizations provide a

variety of excludable and nonexcludable collective goods. Places of worship produce non-

collective goods (depending on the theology), but also produce social capital by fostering

repeated interactions between like-minded individuals. So it is clear that collective-good

provision is a major activity of many nonprofit organizations.

Social investors are happy that nonprofits provide collective goods, but their willingness

to invest depends on the marginal impact of their investment – the additional collective goods

provided when an additional investment is made. Here is where the organizational forms differ.

For-profits have the incentive and ability to reduce the marginal impact of a donation to zero

(Hansmann, 1980). The reason – although investors can observe the amount of collective goods,

they cannot observe the amount invested by others and so cannot tell whether their own spending

has increased total spending on collective goods. For example, suppose 100 investors each

donate $1,000 for a total of $100,000. As long as the for-profit provides more than $1,000 worth

of collective goods, donors will not be able to tell if they have been shortchanged. If the firm

spends, say, $5,000 on collective goods it will enjoy profits of $95,000 and distribute these

7

profits as dividends to shareholders. If one investor doubled their donation, or if a 101 donorst

observed for-profit spending before donating, then added $1,000 to the pool, the for-profit has

the ability and incentive to continue spending $5,000 on collective goods. Neither investor

would know they had been shortchanged, because perhaps some other investor withdrew $1,000.

This illustrates what is now called “marginal-impact contract failure,” and shows why social

investors that want to increase private provision of collective goods should not invest in for-

profits (Bilodeau and Slivinski, 1998). 4

The nonprofit form can be thought of as a standardized contract – a nondistribution

constraint – that is enforced collectively by an entity (usually the State) that can observe total

investments (Krashinsky, 1986). Hansmann argued that nonprofits are likely to avoid marginal-

impact contract failure because there are no shareholders profiting from organizational

opportunism.

Contract Failure in Commercial Markets

Marginal-impact contract failure explains why for-profits do not provide nonexcludable

collective goods, why donors do not give to for-profits, and why donative nonprofits are superior

vehicles when social investors wish to increase the provision of collective goods. But Hansmann

(1980) developed a second kind of contract failure that is appropriate for commercial nonprofits

– those that receive the bulk of their revenues from sales of goods and services like hospitals and

If there is a single social investor contracting with the for-profit to provide a large4

quantity of the collective good, she can do so successfully. Thus, government contracts with for-profits can induce them to produce collective goods. Marginal-impact contract failure occurswhen there are many donors making smaller gifts and the for-profit spends an amount just largerthan the largest individual gift.

8

private universities. Goods and services can be thought of as consisting of a bundle of attributes.

Some attributes are easy for consumers to observe (e.g., whether surgery was performed) but

others are hard or impossible to observe (e.g., whether adverse surgical outcomes were due to the

doctor taking shortcuts). Contract provisions regarding unobservable attributes (or things that the

buyer and seller can observe but the legal authorities cannot) are unenforceable.

For-profits have an incentive to shortchange the consumer on costly attributes that are

difficult to observe and can succeed in doing so as long as the attributes are not third-party

verifiable. “Ethical” for-profits that choose to deliver the promised level of quality face5

shareholder pressures to deliver the kind of profits that other (short-changing) for-profits realize.

As a result, the quality of service delivered by for-profits is lower (in its unobservable attributes)

than that contracted for, another kind of contract failure. In contrast, although nonprofits offer

low-quality outputs on occasion (the soup at soup kitchens rarely wins Michelin awards), they

deliver the quality that was promised. Theory does not predict which sector will deliver higher-

quality goods and services, only that for-profit quality will be lower than represented.

Contract failure in commercial markets results from firms trying to cut costs in ways they

can get away with in order to keep current and prospective owners happy. Nonprofits cannot

distribute any profits to those in control and so have less incentive to shortchange the social

investor. That is the theory, anyway. But whether contract failures are remedied by the nonprofit

form is controversial. As I summarize in more detail elsewhere (Steinberg 2006),

nondistribution removes one incentive but not all incentives for delivering less than the promised

When attributes are hard, but not impossible, to observe for-profits behave better. Then5

opportunistic firms suffer reputational losses, ethical firms can offer meaningful guarantees,consumer rating agencies can inform consumers, and governments can directly regulate quality.

9

quantity of collective goods or quality of commercial goods. Further, it is difficult to enforce

nondistribution, trustworthiness is not the same thing as trust, and many consumers don’t even

know whether the hospital, nursing home, or child day-care center they patronize is nonprofit,

for-profit, or governmental. On the other hand, religious affiliation produces trustworthiness and

trust (but not always), and religiously-affiliated organizations are almost always nonprofit. So

the consumer doesn’t have to know whether an organization is nonprofit to benefit from reduced

contract failure, as long as the choice is made based on characteristics that are absent from for-

profits. Nonprofits are less likely to take advantage of consumers and donors because they are

governed by a volunteer board, have unpaid observers on site (volunteers) that would quit if they

saw any devious behavior, have a charitable mission and culture, pay lower salaries (but not

always) and so attract workers and managers who receive part of their compensation in form of

warm glow from helping to advance the organizational mission.

With strong arguments on both sides of the debate, it becomes an empirical question

whether on balance nonprofits are better than for-profits with respect to contract failure.

Unfortunately, there are no direct tests for contract failure because we are dealing with

unobservables. If the consumer, regulator, and courts cannot observe whether the promised

quality is delivered, how can the researcher? But many indirect tests have been conducted,

summarized in Steinberg (2006), and my personal opinion after reading the studies is that the

average nonprofit is more trustworthy than the average for-profit in markets for goods and

services where contract failure is expected to be important. Thus, social investors should donate

to nonprofits rather than buying shares of for-profits when the target industry produces goods and

services with important non-observable attributes.

10

Merit Goods

Merit goods are goods and services that consumers undervalue according to some

external standard. This is an explicitly paternalistic matter, and controversy can be expected any

time society overrules consumer sovereignty. Sometimes the State provides merit goods, by

banning the sale of large sugary drinks (changing the consumption of those who undervalue

healthy eating), ticketing drivers who don’t strap in (for those who undervalue their own safety),

and imposing prohibitive taxes on alcohol and tobacco (for those who undervalue their own

health and the well-being of others). Sometimes the State provides merit goods by taxing

citizens to pay for free distribution. In both cases, the State exercises its coercive powers to

address the issue, and therein lies the controversy: there is no consensus regarding whether this

constitutes a legitimate exercise of coercive power.

Nonprofits also provide merit goods, through social marketing campaigns,

proselytization, and price subsidies (including free distribution). Social marketing is about

sending messages that change the choices people make in such matters as unprotected sex,

smoking, drinking and driving, and child raising. Some religions seek to change not only the

behavior but also the beliefs of infidels, converting them to faithful followers on the path of true

righteousness. Nonprofit symphonies offer free performances for schoolchildren, hoping to

convert them to lifelong lovers of classical music. Unlike State provision, nonprofit provision is

voluntary (social marketing is prescriptive, not the law of the land) and often financed voluntarily

though donations and sales of goods and services (though tax dollars certainly flow to nonprofits

in the form of grants and, via transfer of burden to other taxpayers, via tax abatements and

exemptions).

11

In some cases, merit goods are supported by social investments in for-profits. By

investing in a company that produces a better nicotine substitute, social investors help to reduce

smoking. The quest for profits is advanced through marketing efforts to persuade people to quit

smoking. In other cases, even a well-motivated for-profit will be ineffective because the for-

profit label creates doubts about the truthfulness and sincerity of social marketing efforts. James

and Rose-Ackerman (1986, p. 53) make this point in a slightly different context:

“[T]he nonprofit form was chosen by the founders because their main objectivewas often not compatible with profit-maximizing behavior or even with theappearance of profit-maximizing behavior. ... For proselytizing reasons, they maywish to present an image of service and dedication, rather than appear “self-serving” and profit seeking. The nondistribution constraint conveys this image, atlittle cost to an organization which did not intend to make large profits anyway.”

Redistribution

Sometimes the goal of profit maximization aligns well with redistributive efforts. Thus,

both for-profit and nonprofit daycare centers for children offer sliding-scale fees; both for-profit

and nonprofit universities offer financial aid. Charging a single price neither helps the poor nor

maximizes profits, so both sectors employ price discrimination, which means that different

customers pay different prices. Price discrimination aids profits because organizations no longer

have to decide between a high margin and low volume price versus a low margin and high

volume one – they can do both. Price discrimination helps the poor, who are offered an

affordable price while organizations collect sufficient revenues from their higher-priced sales to

cover their costs.

The motivation is different, but the pattern of prices is similar for nonprofit organizations

that charge a higher price to some in order to have the resources to subsidize the price for others.

12

However, there are at least two important differences in the pattern of prices between for-profit

and nonprofit price discriminators. First, they differ in the “floor,” the lowest price charged to

any customers. For-profit floors must be higher than the cost of extending service to the last

consumer (marginal cost); otherwise the last consumer would subtract from firm profits.

Therefore, only those customers who can cover the cost of service will be offered prices they are

willing to pay. Nonprofit floors will be lower than marginal cost, and may set at a price of zero,

if the mission-target population requires subsidization. Second, for-profits set prices that

maximize profits. In the limit, when for-profits know each consumer’s maximum willingness-to-

pay, individualized prices are set at each consumer’s maximum. In contrast, because nonprofits

use price discrimination as a means to an end (subsidizing mission-target populations), the

nonprofit price schedule is lower unless the need to subsidize the target population is sufficiently

large and regarded as paramount in organizational decisionmaking (Steinberg and Weisbrod.

1998; 2005). Finally, the two types of organizations differ in their response to increased social

investment: for-profits continue to exploit price discrimination to the maximum feasible extent

whereas nonprofits reduce the prices charged to high willingness-to-pay customers, reduce the

price further for low willingness-to-pay customers, and increase the share of low willingness-to-

pay customers allowed to purchase the good at a subsidized price (Steinberg and Weisbrod,

2005; Starke, 2012).

The Case for Social Investment in For-profit Organizations

There are a few for-profits operated exclusively as social enterprises. More commonly,

for-profits practice “corporate social responsibility (CSR),” so that they combine social enterprise

13

with an ordinary private venture. Kitzmueller and Shimshack (2012, p. 53) survey various

definitions of CSR, concluding: “In essence, CSR is corporate social or environmental behavior

that goes beyond the legal or regulatory requirements of the relevant market(s) and/or

economy(s).” Like my definition of social enterprise, these authors allow exceptions, as the

authors continue (p. 54): “[I]n order to capture its complete economic relevance, this view

emphasizes that CSR can be market driven or “strategic” as opposed to McWilliams and Siegel

(2001), who equate CSR only with social or environmental performance “beyond market forces.”

... CSR may be strategic but need not be.” Although the literature on CSR has not stressed the

point, much of the analysis applies equally well to for-profits exclusively or partly engaged in

social enterprise.

Because I define social enterprise as encompassing things not ordinarily provided by

markets, this ought to be an easy section to write. For-profits do not ordinarily provide

nonexcludable collective goods, overly restrict access to excludable collective goods, fail to

deliver the promised quality in cases of contract failure, cannot effectively provide certain merit

goods, and limit their redistributive efforts to those that add to profits. At least some for-profit

owners want maximal profit distributions; others are constrained in their departures from profit

maximization by the market for control. So it doesn’t matter whether for-profits are more

efficient and nimble because they don’t engage in social enterprise.

But there are exceptions, cases in which for-profits provide collective goods and

otherwise depart from the standard (neoclassical) model of for-profit behavior. In this section, I

first describe the potential advantages of for-profit social enterprise. Then I show when and

where social enterprise emerges in for-profit organizations, either because social production

14

adds to corporate profits or because the market for control allows departures from profit

maximization. I follow that discussion with one regarding limitations of for-profit social

enterprise. The same forces that lead for-profits to engage in social enterprise erode the

efficiency and capital advantages of the for-profit form. I conclude with two case studies.

The For-Profit Advantage

Since Adam Smith, it has been understood that for-profits are sometimes guided, as if by

an invisible hand, towards the common good. Progress since then has made Adam Smith’s

intuition precise as the First Fundamental Theorem of Welfare Economics, which for our

purposes translates to the following: When all goods are produced and traded in perfectly

competitive markets by for-profit firms under conditions of symmetric information, equilibrium

is socially efficient. The impact of this theorem is often exaggerated, so we expand on each term.

First, by social efficiency (also known as pareto optimality), we judge the description of the

entire economy – how much of each good is produced, using what methods and resources, and

how total production is divided among consumers. An economy is socially efficient if there is

no alternative economy that is unambiguously better, in the judgment of consumers. Put another

way, starting from an efficient economy there is no way to make any individual better off without

making at least one other individual worse off in their own estimation. There are no merit goods

allowed in socially-efficient equilibria because consumer sovereignty is fully respected.

The theorem is in “if-then” form. There is no presumption that all goods are traded in the

real world. In particular, nonexcludable collective goods cannot be traded, and side-effects of

market transactions (externalities) are not separately traded in most cases. There is no

15

presumption that information is symmetrically distributed – there is contract failure. Finally,

there are many alternative efficient economies, differing in their distribution of income, and the

theorem offers no reason to suspect that current ownership of resources will lead to a fair or just

distribution of income. The advantages of nonprofits come from each of these cases not covered

by the theorem.

The First Fundamental Theorem offers many insights that do carry over to the real world,

however. Profit-maximizing firms produce at the lowest possible cost (productive efficiency),

seek to innovate, and pay attention to consumer preferences (allocative efficiency). This all

works very well for private goods, whose characteristics are observable, in competitive

industries. Now, my definition of social enterprise rules out such cases, but perhaps productive

and allocative efficiency carry over to for-profit social enterprise when social investment is

needed.

Organizations need capital to grow and adapt to changing preferences. For-profit firms

get the financial capital they need from investments by the founder, initial public offerings, and

debt. Capital is plentiful (in normal times) and available for all projects expected to generate

profits. But financial investors care little about activities that are not rewarded by the market, so

capital is scarce for social enterprise. This is thought by many to be a particular problem of

nonprofits and one that for-profits address, but I argue here that nonprofits have access to more

capital than commonly thought, and later that for-profit sources of capital will dry up when used

to finance social enterprise.

Hansmann (1981) made the case for the for-profit advantage in access to capital.

Nondistribution of profits rules out traditional equity capital – the sale of full-ownership shares.

16

Absent risk-sharing by shareholders, banks and other financial institutions demand higher

interest rates on loans to nonprofits. Thus, it is asserted, nonprofits are short of capital, occupy

too small a share of the economy, and grow too slowly. But nonprofits have sources of capital

not available to for-profits, chiefly donations of time and money, but also, depending on the

jurisdiction, special subsidies or the right to benefit from tax-exempt bonds. In addition,

Hansmann argues, the difference between nonprofit and for-profit entity taxation in many

jurisdictions allows nonprofits to grow more quickly through retained earnings.

Capital flows freely across for-profit organizations in search of the largest financial

returns. In contrast, capital cannot easily flow out of nonprofits, due to asset-lock practices that

prohibit distribution to owners even when nonprofits are terminated or converted to for-profits.

Instead, nonprofit assets must be sold at fair market value, with proceeds dedicated to some other

nonprofit organization whose mission is as close as possible to the original organization’s

mission. Capital does not flow across nonprofit organizations in search of the highest social or

financial returns, and this creates inefficiencies (Hansmann, Kessler, and McClellan, 2003).

Profitable Social Enterprise

The literature on CSR resulting from market forces finds four ways in which corporate

philanthropy or direct social investment adds to the bottom line: enhanced revenue, reduced cost,

increased investment, and reduced public regulation. Consumers will not pay more to buy

identical products from profit maximizers or social investors, so the enhanced revenue models

identify a reason why consumers do not regard the latter as a perfect substitute for the former.

Corporate philanthropy models (e.g., Clotfelter, 1985; Navarro, 1988; Webb and Farmer, 1996)

17

argue that a company’s image affects willingness to pay for its products. Highly-visible

contributions to popular and noncontroversial charities improve that corporate image, leading to

a higher equilibrium price.

In contrast, recent CSR models such as Bagnoli and Watts (2003), Kotchen (2006), and

Besley and Ghatak (2007) consider CSR as the joint production of a private good and a collective

one that serves to differentiate for-profit products. For example, when manufacturers choose a

less-polluting technology, they jointly produce the private good they sell and the collective good

of cleaner air and water. Coffee growers obtaining Fair Trade certification simultaneously

produce a private buzz and collective benefits for those who care about working conditions and

pay in the industry. Or light bulbs manufactured by persons with disabilities produce both

glowing orbs and warm glow. Consumers are willing to pay a bit more for private goods

accompanied by collective ones. As long as the higher price covers the added costs stemming

from joint production, CSR is a profitable activity. Another way CSR adds to revenues is by

reducing the likelihood of boycotts by activists concerned with social performance (Baron,

2009).

Economies of joint production result in CSR by for-profit firms. But these economies do

not provide for-profits with a comparative advantage over nonprofits in most cases. Indeed, the6

joint production model was first used to explain the behavior of commercial nonprofits seeking

Posner and Malani (2007) point out that nonprofit statutes typically restrict nonprofit6

status to organizations in specified industries. Sometimes, there are economies of scope thatmake it desirable to produce social outcomes jointly in an industry not on the list of permittednonprofit uses. In such cases, for-profits have the comparative advantage, they argue. But theyexaggerate the point. In the U.S.A, nonprofits can earn “unrelated business income,”paying a taxsimilar to that paid by for-profits. As long as this unrelated business income does not exceedpoorly defined limits, the two sectors are on an equal footing here.

18

to cross-subsidize their nonmarket activities (Posnett and Sandler, 1986).

Corporate philanthropy and other social investment cuts costs in several ways. First,

these activities can reduce recruiting and turnover costs, particularly for key executives that

prefer to work for a company that makes social investments and is seen to be virtuous. Second,

as pointed out on the revenue side above, CSR cuts costs when there are economies of scope

between private and collective goods (where producing outputs together is cheaper than

separately). Third, CSR can serve as a screening device, attracting morally-motivated workers

that will work for the interests of the organization without costly incentives and oversight

(Brekke and Nyborg, 2004). Again, for-profits have no comparative advantage over nonprofits

in these sources of cost reduction.

Social investors buy shares of both profit-maximizing and CSR firms and also make

direct donations to nonprofits in the model of Graff Zivin and Small (2005) and subsequent

extensions by Baron (2007). Profit-maximizing firms distribute all their profits as dividends,

whereas CSR firms divert a share of profits to charities. Investors care about their consumption

and their direct and indirect (via ownership of CSR firms) contributions to charity. But investors

differ from each other in the relative value they place on a dollar donated directly vs. indirectly.

This leads to diversity in shareholder portfolios, with some investing only in profit-maximizing

firms and making personal donations and others buying shares in both kinds of organizations

instead of donating.

Expected spending on CSR does not harm investors in these models because the price of

a share of stock reflects the expected stream of future dividend payments. The CSR firm pays

lower dividends, but it costs less to obtain the rights to those dividends. Baron quoted John

19

Mackey (2005), the CEO of Whole Foods, and I reproduce that here:

I believe the entrepreneurs, not the current investors in a company’s stock,have the right and responsibility to define the purpose of the company. It is theentrepreneurs who create a company, ... and who negotiate the terms of trade withall of the voluntarily cooperating stakeholders – including the investors. At WholeFoods we “hired” our original investors. They didn’t hire us.

We first announced that we would donate 5% of the company’s net profitto philanthropy when we drafted our mission statement. ... Our policy ... predatesour IPO by 7 years. All seven of the private investors at the time we created thepolicy voted for it when they served on our board of directors. ... How can WholeFoods’ philanthropy be “theft” from the current investors if the original owners ofthe company unanimously approved the policy and all subsequent investors madetheir investments after the policy was in effect and well publicized?

When future social spending levels are known by investors, the costs of social spending are

entirely borne by founding entrepreneurs. Firms that contemplate reduced profits as a result of

CSR are not formed unless there are entrepreneurs willing to bear that burden. Social investors

will buy shares, but organizations do not become responsible in order to attract more investment.

The most surprisingly result of these models is that total social spending is gnerally

higher as a result of corporate social spending despite crowding out of investor donations to

charity caused by the CSR activities of the firms they invest in. The discount price of CSR

shares causes an increase in share ownership, resulting in more giving via share ownership than

would occur if the only option were direct investor gifts to charity. However, social spending

generally remains sub-optimal in the sense that average well-being would be higher if more were

spent on collective goods.

Baron (2009) combines revenue and investment factors in a single model and concludes

(as summarized in Kitzmueller and Shimshack, 2012, p. 67):

“Equilibrium levels of CSR will vary across types and depend on the degree of

20

substitutability between the various social contribution channels, i.e., invest,consume, donate or support an activist.”

Finally, improvements in corporate image stemming from corporate philanthropy can

forestall costly regulatory and legislative initiatives. The interested reader is referred to the

formal analyses in Maxwell, Lyon and Hackett (2000) and Calveras, Ganuza, and Llobet (2007).

Social Investment and the Market for Control

For-profit ownership rights are for sale. In a stock corporation, shares are bought and

sold and the largest shareholders have control. Generally, a shareholder gains a controlling

interest with far less than half the outstanding shares; majority ownership is only required when

other shareholders band together as one. Privately-held for-profits also sell ownership rights

through the transfer of shares, but there is less transparency as shares are neither traded on a

public exchange nor subject to the reporting and regulatory constraints commonly placed on

publicly-held firms. It is harder to sell ownership rights in other forms of business enterprise

(like partnerships and sole proprietorships in the U.S.), but all the organizational assets can be

sold (including the name and goodwill) to a new entity.

In the canonical model of for-profits, the market for control limits behaviors that depart

from profit-maximization. For example, Friedman (1970) characterizes CSR as the spending of

stockholders’ money by corporate executives and asks “Will not the stockholders fire him?

(Either the present ones or those who take over when his actions in the name of social

responsibility have reduced the corporation’s profits and the price of its stock.)” In the canonical

model, CSR share prices are lower than they would be if the firm maximized its profits, and this

21

difference in price motivates takeover bidders who gain control, return the firm to profit

maximization, and then sell at a profit. More generally, Baron (2007, p. 705) concludes that “To

be sustainable, CSR firms must be protected from the market for control.”

And, in more realistic models, firms are protected from the market for control, to varying

extents, by laws, regulations, diversity of investor preferences, and informational asymmetries. If

investors as principals want firms to maximize their profit distribution, managers as agents have

other objectives. This principal/agent problem can be reduced by various costly measures, but

not eliminated. At least some corporate philanthropy is motivated by the private and social

interests of managers who advance their social status or further social welfare through the

particular gifts they make. Most observers feel that this kind of philanthropy is shrinking when

compared to profit-oriented “strategic philanthropy” (e.g., Porter and Kramer, 2002). But to the

extent that shareholders cannot control management behavior, for-profits are free to pursue

alternative missions, including social enterprise but also including cronyism, excessive

compensation, and other less-desirable behaviors.

The rights of shareholders vary greatly from place to place and over time. Place enough

constraints on shareholders and they will fail to discipline managerial departures from profit

maximization, either because they cannot secure the necessary information or they cannot

enforce their will on management. In addition, when ownership is relatively dispersed,

individual shareholders would rather free-ride or simply buy stock in another firm than invest

their time and attention to improving the quality of management. For-profits protect themselves

from the threat of takeover bids with a variety of strategies like poison pills and golden

parachutes. The laws regulating such practices vary over time and place and enforcement varies

22

as well. Wikipedia’s (2012) discussion of “Takeovers” bemoans the lack of citations but has

concluded since October 2006 that:

Corporate takeovers occur frequently in the United States, Canada, UnitedKingdom, France and Spain. They happen only occasionally in Italy becauselarger shareholders (typically controlling families) often have special board votingprivileges designed to keep them in control. They do not happen often in Germanybecause of the dual board structure, nor in Japan because companies haveinterlocking sets of ownerships known as keiretsu, nor in the People's Republic ofChina because the state majority owns most publicly-listed companies.

Takeovers do not occur if the owners value CSR more than the maximum financial gains

(difference in share prices) that would be offered by takeover bidders. In summary, the market

for control has less influence on for-profit behavior when the number of owners is small, the firm

is privately held, and the current owners have chosen to buy share because of the high value they

place on CSR.

Limitations on For-Profit Social Enterprise

Having made the best case I can for for-profit social enterprise, I conclude that the usual

advantages of for-profits disappear when they engage in social enterprise, and in any case for-

profit social enterprise is limited in volume and scope. Because of the diversity of definitions

and a paucity of data, it is difficult to assess the true extent of for-profit social enterprise. One

element of this is revealed by the volume of private social investments. A report by the Social

Investment Forum Foundation (2010, Executive Summary p. 8) estimated that “The total assets

managed under policies that explicitly incorporate ESG criteria into investment analysis

and portfolio construction (ESG assets) are valued at $2.51 trillion.” This compares with about

23

$25 trillion in managed assets in the U.S.A. Corporations devote a fraction of that sum to annual

donations – $14.1 billion in cash and in-kind gifts in 2009 (Giving USA 2010). This is only

about 4% of total giving of $304 billion. The Conference Board survey of major corporations

found U.S. median contributions, as a percentage of pretax income, were only 1.55% in 2008.

Surveyed corporations gave an additional 2.59 billion abroad in 2008, mostly in-kind (Cavicchio

and Torok, 2009). A survey conducted by The Social Investment Consultancy (2010) finds that

“British business gives an estimated £1.4 billion a year in revenue and support each year to good

causes.”

Corporate philanthropy is also limited to noncontroversial causes. Those who support

controversial causes risk boycotts and bad public relations, deterring some from giving and

penalizing those who continue to give (Himmelstein, 1997). Thus, it is not surprising that

Himmelstein concluded (as summarized in Galaskiewicz and Colman, 2006, p. 184):

[G]iving officers in particular had a strong commitment to do somethinggenuinely worthwhile for the communities in which their firms operated. Yet,doing good was difficult to defend in companies that were under attack bydisgruntled shareholders, embroiled in cutthroat competition, or vulnerable tocrises beyond their control. Because the function often did not directly contributeto the “bottom line,” to survive it had to have the support of the CEO or chairmanof the board or it had to speak to the strategic interests of the firm. Yet, to ensureits integrity the giving program had to guard against becoming a “plaything” ofsenior executives or an arm of the marketing/personnel/public relationsdepartments. This is often a difficult tightrope to walk.

Social expenditures are encouraged when consumers are willing to pay a higher price for

purchases from socially-responsible firms. But in the joint production model of Bagnoli and

Watts (2003) the level of private provision of the collective good varies inversely with the

competitiveness of the private-good market.

24

Investors are willing to buy shares in CSR firms. But as Graff Zivin and Small (2005)

and Baron (2007) show, corporate social expenditures come partly or entirely at the expense of

individual donations. In the special case where all investors view corporate spending as a perfect

substitute for individual giving, total social expenditures are unaffected by CSR. Every increase

in corporate spending causes an equal and opposite decrease in personal donations. Further,

because the costs of anticipated CSR are borne entirely by the founding entrepreneur in these

models, the extent of CSR is limited by the scarcity of entrepreneurs willing to make the

necessary sacrifice.

Although the market for control is far less effective in preventing departures from profit-

maximizing behaviors than suggested by the canonical model, it is never entirely absent and so

there are always questions regarding the sustainability of CSR spending. Current owners may

have all bought-in to a social mission as Mackey (2005) suggested, but current owners die,

preferences change over time, and those placing a lower value on CSR have the incentive to

mimic their more socially-responsible fellow investors. Rodgers (2005) commented:

Mackey spouts nonsense about how his company hired his originalinvestors, not vice versa. If Whole Foods ever falls on persistent hardtimes—perhaps when the Luddites are no longer able to hold back the geneticfood revolution using junk science and fear—he will quickly find out who hashired whom, as his investors fire him.

Even when the owners are somewhat in agreement, managers face the difficult problem

described by Yunus (2010):

When you mix profit and social benefit and say that your company willpursue both goals, you are making life complicated for the chief executive officer.His thinking process gets clouded. He does not see clearly. In a particular situationwhere profit and social benefit need to be balanced, which way should the scalesbe tipped? ... What about in times of economic stress, such as a recession—is it all

25

right to eliminate social benefits altogether in hopes of helping the company tosurvive? ... The idea of a "mixed" company offers no clear guidance on questionslike these. In practice, profit tends to win out in struggles of this kind. Most oftenthe CEO will lean—perhaps unconsciously—in favor of profit and exaggerate thesocial benefits being created. ... [T]he social goals will gradually fade inimportance while the need to make money becomes more and more deeplyingrained in the company's culture.

The problem is magnified when for-profit social enterprises change owners. Mac Cormac et al.

(2007, 97) conclude that “Many socially oriented for-profits find that their social mission is

dependent on founders’ fervor, and when founds retire or sell, their social legacy is often lost as

more traditional owners take over.” Katz and Page (2010, p. 96) view the problem “as an inter-

temporal agency problem between an idealist founder (principal) and his older and perhaps more

acquisitive self (agent).

Social expenditures by for-profits are only sustainable when the market for control is

weak or absent. But this absence allows for-profits to sustain inefficient behaviors, so the

efficiency advantage of for-profits is reduced or lost entirely when these firms engage in social

enterprise.

For-profit access to equity capital is plentiful for profit-maximizing activities. But

investor’s purchase of CSR shares is, in the various models examined, the joint purchase of a

stream of future dividends and donations. Although Baron (2007) showed that this arrangement

can, under specified circumstances, increase total social investment relative to a pure private

donations equilibrium, the increase is limited. Therefore, for-profit access to capital for social

investment is not as bountiful as it is for profit-seeking investments.

None of the CSR models discussed above include informational asymmetry between

26

purportedly socially-engaged for-profits and consumers, donors, workers, and investors. For-

profits have the same incentives here as elsewhere, and I believe contract failure is the most

common result. It pays to exaggerate the degree of corporate social expenditures, as gullible

consumers would overpay and gullible workers would accept lower financial compensation.

Lyon and Maxwell (forthcoming) model one aspect of informational asymmetry which they call

“greenwash,” the selective disclosure of positive but not negative information about a firm’s

environmental performance. There are private social-responsibility rating services and laws

against fraud, but for collective goods marginal-impact contract failure is nonetheless likely.

I’ve experienced a likely example when I was told that a certain for-profit would

dedicate a share of their proceeds to a specified charity. I asked, and was told there was a cap – a

maximum amount that would be given away. Interested in the marginal impact of my purchase, I

then asked whether the ceiling had been reached or would be likely to be reached absent my

purchase. When the ceiling binds, my purchase does not help the charity. The marketer was

unable to enlighten me on this point. Perhaps the for-profit had inadequate internal tracking or

some other innocent explanation applies, but I suspect this was an attempt to exaggerate social

expenditures in order to enhance profits.

Case Studies7

Newman’s Own, Inc. was founded by Paul Newman and privately held. Since its

beginning, Newman’s has successfully donated all after-tax profits to the Newman’s Own

For another case study, see Page and Katz (2010). They conclude that the sale of Ben7

and Jerry’s Homemade Inc. to Unilever by the founders was not compelled by corporate law, anddiscuss what parts of the philanthropic behavior and culture survived the change in ownership.

27

Foundation, a grant-making nonprofit. The founder strongly wanted all profits donated to

charity, and got his way, a policy that continues after his death with donations exceeding $350

million (www.newmansownfoundation.org, accessed June 30, 2012). How did Mr. Newman

succeed in the face of the market for control in sustaining his mission for so long? The key is

that Newman was the sole initial investor. He could have taken a distribution of profits at any

time, but chose not to do so because he valued the social mission more than the potential profits

he could receive as a distribution. Of course, the maximum takeover bid that could be advanced

would be no greater than the potential distributable profits, so a mutually-beneficial transfer of

ownership did not and could not take place. That is why Brakman Reiser (2009, p. 2451)

concluded that “social enterprise companies are often small and controlled by owners who have a

personal commitment to their social goals. A few have begun that way, but grew quite large as a

result of their success.”

If Newman required investments exceeding his own ability to found the for-profit,

ownership would be split among several investors. Perhaps Newman could have found a

sufficient number of investors that shared his vision, and the for-profit would continue to operate

charitably without pressure from the market for control. But if he guessed wrong, if he found

investors that claimed to share his social mission but acted otherwise when the firm became

profitable, then the market for control would make the social mission unsustainable. Prejudging

investor motivations is difficult because of a demand-revelation problem: greedy investors would

pretend to share the social mission in order to vote for profit distribution at a later time. Even if

the original set of investors were sufficiently committed to escape the market for control,

ownership shares would be transferred upon death and at any time there is no guarantee that

28

those supporting Newman’s original mission will continue to enjoy a controlling stake in the

firm. 8

Newman did not face the initial “demand revelation” problem because he needed no other

investors. But he did die, prompting me to write the following to Roberta Pearson, who responds

to consumer mail sent to newmansown.com (sent May 7, 2010):

I am a Professor of Economics and Philanthropic Studies. The question ofwhether for-profit firms can sustainably devote 100% of profits to charity evenafter the death of a charismatic founder has come up in my research. ... I cannotfind out from your web site whether Newman's Own, Inc. is publicly traded. Is it? Are there particular ways in which you try to assure that future owners will shareNewman's mission? Please note I understand that the Newman's Own Foundationwill continue to work as it has – I am asking about ways to insure that thefor-profit Newman's Own Inc. will continue to devote 100% of profits to thefoundation or directly to charity.

Ms. Pearson responded (May 12, 2010):

Newman's Own, Inc., a for-profit company was privately held by PaulNewman. Mr. Newman donated all his royalties and after taxed profits from hisfood company to the charities of his choice. When Mr. Newman became ill heformed the Newman's Own Foundation and the Foundation now owns Newman'sOwn, Inc. Now all royalties and after taxed profits from Newman's Own, Inc., goover to the Foundation and their board members disburse those monies tocharitable organizations.

Aha! By making the for-profit Corporation into a wholly-owned subsidiary of the nonprofit

Foundation, we have a hybrid that cannot legally distribute profits. The same could be

accomplished by merging the Corporation into the Foundation, which would then pay the

Brakman Reiser (2010) makes a similar point (p. 645): “Smaller, closely-held8

corporations and unincorporated forms offer somewhat greater room for blended objectives. ...Still, the appetites and desires of owners can shift, and even closely-held and unincorporatedbusiness forms are business forms. If a set of owners come, over time, to favor profits oversocial goals, it will be hard to constrain them from voting their preferences. If enough ownerscome to this position, their votes can dominate their fellows and change the course of the entity.”

29

“unrelated business income tax”that applies to nonprofits rather than the for-profit corporate

income tax. In other words, for all intents and purposes, the Corporation was converted to a

nonprofit, with all the comparative advantages and disadvantages of the form.9

I conclude that for-profit social ventures are possible and will endure for a while if the

founder has personal access to the requisite initial capital and remains committed to the social

mission. But founders do not endure forever; eventually like all mortals they founder. Then,

unless the venture is converted or effectively converted into a nonprofit form, there will be a

growing risk that the organization will revert to traditional profit-maximizing behavior with

distributions to shareholders. The larger the number of required investors, the smaller the chance

that for-profit social ventures will be formed and if formed, the shorter the time they will endure.

Google.org (DotOrg) was created as a for-profit philanthropy by Google Inc. with an

initial donation of 1% of Google’s equity and the transfer of management over grantmaking by

the earlier, nonprofit, Google Foundation. In addition, the Corporation donates employee time

and money to DotOrg. Like a traditional foundation, DotOrg makes grants, but has shifted most

of its efforts towards developing engineering solutions for urgent world problems (Google.org,

2012). Unlike a traditional foundation, there is no asset lock – upon dissolution, assets would

revert to the owner, Google Inc. This feature deters donations from outsiders. Unlike Google

Conceivably the Foundation could sell its subsidiary and then there would be no9

guarantee that the Corporation would continue to donate all net profits to charity. But as anonprofit, they would only do so at fair-market value or above, so they would receive a sum atleast equal to the present value of the stream of expected future Corporate net profits. In furthercorrespondence with Ms. Pearson the same day, I was gratified to learn that “There are no plansto divert from Mr. Newman's vision.”

30

Inc., DotOrg does not issue shares to outsiders, so its access to capital is restricted by Google’s

admittedly deep pockets.

DotOrg is unique in taking this approach. They chose to forego the many advantages of

the nonprofit form (including tax exemption and the right to receive tax-deductible donations)

because of the ways nonprofits are regulated in the U.S. As a for-profit, they escape restrictions

on investing in other for-profits, restrictions on direct access to the resources of Google, Inc., and

restrictions on political activities stemming from federal tax law and state regulation (Brakman

Reiser, 2009).

Because Google, Inc. announced plans to devote 1% of profits to social investment prior

to its initial public offering, Brakman Reiser (2009, p. 2468) concludes: “Any shareholders

disdaining such use of their capital could have looked elsewhere to invest. ... In fact, some

shareholders may have purchased because of Google, Inc.’s philanthropic commitment , and

would want to hold the company to its promises.”

The demand revelation problem has no impact, and there is little pressure from the

market for control because Google, Inc.’s contribution is only 1% of its profits, an amount typical

of traditional corporate donations. DotOrg is visible to the public, contributes to Google, Inc.’s

corporate image, and so enhances marketing and reduces political will to impose costly

regulations or pursue antitrust cases against the parent. Google, Inc.’s executives and key

employees support the investment in social enterprise, so the investment reduces recruiting and

turnover costs. All these factors combine to make Google, Inc. unattractive to hostile takeover

bids that seek to end the investment in social enterprise.

31

Thus, Google.org’s social mission is sustainable, at least for the short and medium term. 10

But it is sustainable because it is like traditional corporate philanthropy, and like traditional

corporate philanthropy, will be limited in size and scope. Perhaps a few firms that do not

currently make corporate donations would be willing to make the kind of reversible donations

embodied in the Google, Inc./Google.org partnership, so the model may increase social

investment a bit, but it will not greatly transform the limits on corporate social investment

stemming from the market for control.

New Hybrid Organizational Forms

Is there a way to combine for-profit access to traditional equity capital markets and

productive efficiency with nonprofit trustworthiness, access to grants and donations, and mission

sustainability? In the U.S., a few such attempts have been made, but as Kelley (2009, p. 3) puts

it:

“A few practitioners have learned to cobble together complex structures ... that

draw on a mix of for-profit and nonprofit forms and doctrines to create legal

scaffolding for hybrid ventures. But those complex structures, which involve

Strom and Helft (2011) detail problems DotOrg has had with key personnel, turnover,10

alleged conflict of interest, and changing aspirations. Approved grants lacking only a finalsignature were cancelled. They cite veteran employees complaining of “how Google’sengineering-centric approach frustrated and limited them.” One anonymous former employeesaid “We concentrated on complicated engineering problems rather than large developmentchallenges. That meant we were creating solutions that were looking for problems rather than theother way around.” Google Inc. donated $184 million in cash and in-kind gifts in 2010, of which15 percent went to DotOrg projects. The problems may just be growing pains, but are consistentwith the inherent conflicts of pursuing a double bottom line.

32

corporations with multiple classes of stock and detailed shareholder agreements,

or the creation of multiple interlocking entities, or the use of delicately drafted

joint venture agreements, tend to be expensive to create, burdensome to maintain,

and, due to their novelty, legally insecure.”

That is why various U.S. state and tribal governments, starting with Vermont, have enacted laws

allowing for the incorporation of Low-Profit Limited Liability Companies (L3Cs) starting in

2008. In the UK, legislation too effect in 2005 providing rules for creating and operating11

another new kind of hybrid organization, Community Investment Corporations (CICs).

Alternative new hybrid forms continue to emerge (e.g., Social Impact Bonds and B Corporations

in the U.S.A.

The most flexible corporate form in the U.S.A. is called a Limited Liability Company

(LLC). This form combines the structural features of a partnership with the limited liability of a

corporation. LLCs have members, instead of shareholders. There can be multiple classes of

members, each with distinct governance rights. Different member classes may require different

financial contributions, and distributions of profit need not be split in proportion to member

investments. LLCs are treated like partnerships for federal tax purposes in that income is taxed

only when received by members. Hybrid organizations for social enterprise can incorporate with

a variety of structures under the LLC designation, but legislation enabling distinct classification

as an L3C was designed to create a “brand name” organization with standard features that

stakeholders would recognize. To qualify as an L3C, the organization must significantly further

Vermont’s statute is 11 Vt. Stat. Ann, tit. 11, §3001(27) (2010). The same or similar11

language was used in the various other states and tribes subsequently adopting L3Cs.

33

one or more charitable or educational purposes (as defined by the federal tax code), not have the

production of income or appreciation of property as a primary purpose, demonstrate that it would

not have been formed but for the company’s relationship to the accomplishment of one or more

charitable or educational purposes, and not include influencing legislation or political

campaigning among the organizational purposes.

Brakman Reiser (2010, p. 628, footnotes omitted) summarizes a tranched membership

structure proposed by advocates of the L3C, designed to promote access to equity capital and

incentivize efficiency while pursuing a sustainable social mission:

An equity tranche of members could be tax-exempt private foundations makingprogram-related investments [PRIs]. Because the PRI regulations specifically barfoundations from contemplating a financial return as a motive for investment, thistranche of members would be given scant or very remote rights to distributions. Amezzanine tranche of individuals or entities could purchase L3C memberships asa type of socially-responsible investment. This tranche of investors would agreeto operating agreement terms that provided them with some access todistributions, but at a rate lower than market return, presumably doing so in returnfor the social or psychic value produced by the entity. The L3C’s operatingagreement could then provide for a market-like return to a senior tranche ofindividuals and entities seeking such returns, presumably doing so in competitionwith other market-rate investment opportunities. The structure of these provisionsmight be more debt-like or equity-like (though if the latter, more like preferredthan common stock), providing either a guaranteed return or a return keyed to theL3C’s profits.

The social mission is enshrined by giving equity tranche members controlling or

exclusive governance rights. This implies that conflicts between mission and profits would be

decided in favor of mission and protects the organization from profit-seeking takeover bids. If,

despite this structure, the social mission of the organization loses primacy, the statutes specify

that the organization reverts to a regular LLC. There is no statutory asset lock preventing the

34

accumulated assets from being distributed to members as a consequence of that reversion.

In the UK, CICs can be formed as a company limited by shares (analogous to U.S. for-

profit) or limited by guarantee (analogous to U.S. nonprofit), with additional requirements and

variations. In general, directors of companies limited by shares have a fiduciary duty to

shareholders but can consider other interests. CICs limited by shares must place these other

interests as primary goals. In general, companies limited by guarantee have one or more eligible

charitable purposes. But if these companies chose to incorporate as a CIC, they lose the tax and

certain regulatory advantages given to charities. Regardless of form, directors must preserve the

organization’s ability to pass the “community interest test,” which specifies that “a reasonable

person might consider [the CIC’s] activities are being carried on for the benefit of the

community” (Regulator of Community Interest Groups, 2009). The CIC must report its

compliance with this test and must “confirm that access to the benefits it provides will not be

confined to an unduly restricted group” (ibid.). Members, who may be shareholders, donors, or

specified others have some governance rights, and have a duty to monitor that the company

continues to meet the community interest test and fully involves the community in its activities

and development.

The Regulator of CICs is empowered to limit dividends for CICs limited by shares. After

an initial problem securing investors, the Regulator eased up and capped dividends at 35% of

distributable profits. In addition, the Regulator enforces an effective asset lock – the

organization cannot sell its assets at less than fair market value except in pursuit of mission or

when transferring assets to another charity or CIC. This lock includes distribution of assets

following dissolution or conversion to a different structure.

35

Other emerging hybrids are described briefly, as the set of recommended alternatives is

fluid and rapidly evolving. The U.S.A. nonprofit B Lab created the label “B-corporation” for for-

profits that meet its certification standards. B-corporation’s web site says

(http://www.bcorporation.net/about):

B Corps, unlike traditional businesses: Meet comprehensive and transparentsocial and environmental performance standards; Meet higher legalaccountability standards; [and] Build business constituency for public policiesthat support sustainable business.

B Lab also successfully lobbied for legislation (passed in seven states, with more pending) to

create a new corporate form, the Benefit Corporation (BC). Like L3C status, BC status is

designed to create a brand name for social enterprise. BCs differ from other for-profits in that the

enabling legislation uniformly and transparently specifies that the fiduciary duties of directors

encompass interests beyond the shareholders. A corporation becomes a BC if approved by two-

thirds of the shareholders. A supermajority is also required for mergers and conversions

whenever the successor organization would not be a BC. Benefit corporations are required to

have a purpose of creating “general public benefit” and are allowed to identify one or more

“specific public benefit” purposes from a nonexhaustive list. BCs must produce an annual report

on its economic, social, and environmental performance, with requirements to insure independent

third-party evaluation of these claims (Benefit Corporation, 2012).

Social Impact Bonds (SIB) are a new concept, combining equity-like and bond-like

tranched investment opportunities for social enterprise (Social Finance, Inc., 2012; Callanan and

Law, 2012). Under this arrangement, government agencies ask consortiums of investors and

donors to bid for a contract between the government and a service provider, for-profit or

36

nonprofit. The contract specifies detailed performance criteria, and the winning consortium (the

bondholders) is responsible for funding and overseeing the project. Risk is transferred from

government and service providers to bondholders because government payments to bondholders

is contingent on the service provider meeting the performance objectives. The first SIB was

issued in the UK in 2010, and Massachusetts issued requests for proposals for SIB grants in

2012.

Evaluating Hybrid Organizations

Advocates of hybrid organization seek the best of both worlds – to combine the

advantages of for-profits and the advantages of nonprofits in a single entity. In this section, I

evaluate this claim. First I look at whether hybrids obtain the efficiency advantage asserted for

for-profits. Then I examine whether they are likely to secure more capital for social enterprise.

Then I ask whether hybrids are likely to avoid contract failure. Finally, I look at mission-

sustainability for hybrids.

Hybrid Efficiency

Hybrids are profit-distributing organizations, so they are thought to provide the efficiency

advantage of the for-profit corporate form. However, extension by analogy is logically dubious,

and I do not find this argument to be persuasive. For-profits are efficient for three reasons: the

owners want efficiency, the market for control eliminates inefficiency, and competition results in

the death of inefficient for-profits. In hybrid organizations, there are various ownership

structures but generally speaking, those investing in order to obtain financial returns (e.g., the

senior tranche in L3Cs) are given little governance power (which rests primarily with the equity

37

tranche in this example). So unlike regular shareholders, who have the choice of voice

(demanding management change) or exit, financial investors have no choice. If they don’t like

the returns they are getting, they invest elsewhere. For this same reason, financially-motivated

takeover bids will not occur.

In addition, asserted for-profit efficiencies and nonprofit inefficiencies have been

exaggerated. Note first that the First Fundamental Theorem claims that equilibrium is efficient.

There is no theoretical reason to think that this equilibium is ever reached in the real world,,

because the underlying fundamental parameters (tastes and productivity) change and we still

don’t fully understand the dynamics of adjustment under perfect competition. Second, the “if”

part of this “if-then” theorem is not true in the real world. Markets are not fully competitive,

some goods are untraded, and there are myriad principle-agent problems impairing efficiency.

Firms employ costly strategies to immunize themselves from takeover (poison pills, golden

parachutes), creating inefficiency and impairing the market for control. Third, efficiency is

attained by the death or restructuring of inefficient for-profits. Unless this death is instantaneous,

at any point in time there will be many for-profits that are “Not dead yet” (Python, 1975),

lowering the average efficiency of for-profits. In practice, corporate follies, particularly by “too

big to fail” firms, make it increasingly hard for me to believe that for-profits are generally

efficient.

Conversely, claims of nonprofit inefficiency are exaggerated and in some cases,

nonprofits are more efficient than for-profits. Money is not the only way to motivate people, and

in areas of social enterprise, money is often not the best way to motivate people. True, nonprofits

are relatively restricted in their ability to use financial incentives to motivate their workforce.

38

But Slivinski (2002) showed that provision of dividends-in-kind, worker enjoyment of the

collective output of the firm, can be the most efficient way to motivate team production by

workers. True, nonprofit directors do not have pecuniary incentives to instill efficiency in the

organizations they control. But waste impairs mission accomplishment, which can powerfully

motivate board members and top managers. True, potential takeover bidders do not search for

nonprofit inefficiencies, but there are many other agents who do. More details, and empirical

evidence on these points are summarized in, e.g. Steinberg (1987; 2006) and Schlesinger and

Gray (2006).

Ironically, hybrids may be efficient because they, like nonprofits, are controlled by those

concerned with the social mission of the organization. But whether we are talking hybrid

organizations or commercial nonprofits, the clash of cultures between those charged to seek

profits and those charged to pursue social objectives is a continuing source of inefficiency.

Absent economies of scope, ongoing fights among stakeholders are costly to mediate and

competing stakeholders devote considerable scarce resources to those fights.

Hybrid Access to Capital

Do hybrids get more capital than for-profits? Advocates assert the leveraging power of

social investment to attract profit-seeking investment. Again, I don’t find this plausible on

theoretical grounds. Some advocates are using different definitions, but because I am defining

social investment as investment in things that are not ordinarily done by markets, the overall rate

39

of return must be below market. In tranched investment structures like L3Cs and SIBs, the12

senior (market return seeking) tranche would need to be cross-subsidized by investors in the

other tranches. The ability to do so is limited, and the desire of social investors to subsidize other

investors is questionable. This is particularly true for SIBs, because profit-seeking investors will13

need a higher rate of return (to compensate for the risk that the bond will not be redeemed by

government if the program fails to meet performance criteria). Private foundations in the U.S.A.

would face additional constraints if the IRS viewed cross-subsidization and risk transfer as a

“jeopardizing investment.”

Advocates for L3Cs believe that additional capital would come in the form of low-

interest loans made by private foundations (Program-Related Investments or PRIs). Indeed the

enabling legislation borrows language from the tax code permitting PRIs that meet specified

criteria. To date, IRS guidance has been unclear, and few private foundations have been willing

to make PRIs (e.g., Kelley, 2009; Doeringer, 2009-10). On April 19, 2012 the IRS proposed

modifications to the Code of Federal Regulations that, if adopted, would reduce the legal

uncertainties and encourage PRIs. (IRS, 2012)..

CICs limit dividend payments, and this restriction reduces their ability to raise equity

capital. Doeringer (2009-10) notes how an early CIC success exhausted its credit lines after it

In cases where for-profit social enterprise is profitable, this does not hold but as such12

opportunities are scarce and do not encompass all areas of desirable social enterprise, mydiscussion here remains relevant.

Absent a formal model, which is for future research, I am hesitant to speculate too much13

on this point. If the limited amount of available cross-subsidization attracted sufficient marketcapital, social investors would be willing to cross-subsidize but I am unsure whether such acondition could exist. Formal modeling would also have to take account of Baron’s (2007)results on the incidence of social investments between entrepreneurs and outside investors.

40

was unable to tap the equity lines due to limited investor interest until it was forced to sell nearly

all its assets to the private sector. He concludes (p. 315):

It remains to be seen whether the dividend restrictions will prohibitively limitCICs’ access to capital, or whether creative social valuation metrics can nurture anactive social-equity market. However, the rapid pace at which entrepreneurscontinue to register CICs shows hope, at least, for the latter.

Hybrid Contract Failure

To what extent do hybrids avoid contract failure? In theory, this depends on the hybrid

structure. Although both L3Cs and CICs are governed primarily by parties whose primary goal

is not profit distribution, profits can be distributed on dissolution or conversion of an L3C. This

provides an incentive for profit-motivated agents to invest in the equity or mezzanine tranche,

pretending to be mission-motivated. These mimics could then vote for dissolution, resulting in

marginal-impact contract failure at least. Regular nonprofits and CICs are constrained by an

asset lock, making attempts to mimic mission-motivated investors less tempting.

Nonprofit trustworthiness is always constrained by the need to secure adequate resources.

In resource-scarce environments, nonprofits may be forced to compromise or go out of

business. Would hybrids become so reliant on market-rate capital that they would be forced to14

concede de facto control to these investors?

Hybrid Sustainability

Space precludes a full discussion of this very complex matter. Hirth (1999) considers a14

model of competition between true nonprofits, for-profits, and for-profits-in-disguise(organizations that secretly distribute profits but represent themselves as nonprofit). He findsthat competition can reduce the trustworthiness of nonprofits or enhance the trustworthiness offor-profits, depending on the quality of enforcement of nondistribution and other factors.

41

In theory, control by mission-oriented parties insures against takeovers or conversions

that would shift the hybrid towards profit-maximization and hence away from social objectives.

That control is weakest for certification hybrids like B-corporations, and weaker for L3Cs than

for hybrids constrained by asset locks. Katz and Page (2010, p. 98) point out that some social

enterprises should not last forever, asking “Under what circumstances should the demise or

acquisition of a for-profit social enterprise be welcomed rather than opposed?” and partly answer

this question by pointing out that “a social enterprise can succeed by demonstrating that it is no

longer necessary and that the market failure the nonprofit undertook to address no longer exists.”

So difficulties in sustaining the organization’s social mission are not always a problem.

Conclusion

In this paper, I argue that social investors should direct the bulk of their investments to

nonprofit organizations. For-profit social enterprise does exist, but has limitations and

drawbacks in certain settings, particularly with respect to contract failure. But there are also

drawbacks to nonprofits.

The world is too complex to be characterized by black and white, and adding current

hybrids as shades of grey does not complete the characterization. There are a variety of emerging

legal structures that attempt to combine the advantages of nonprofits with those of for-profits.

Some observers are quite skeptical that this can ever work. In her article on failed hybrids, New

York Times reporter Strom concluded (2010):

Like Dr. Dolittle’s pushmi-pullyu, the animal that had trouble moving because its

42

two heads could not agree on a single direction, the hybrid model for nonprofits isproving problematic. On occasion, the need to generate returns for investorsoverwhelms the social mission. In other cases, the business falters altogether andcannot support the nonprofit.

Although it is worthwhile to think about additional hybrid structures that might combine the best

of both sectors, there is always the danger of combining the worst of both sectors. In conclusion,

I endorse the conclusion reached by Brakman Reiser (2010, pp. 654-5) and cannot state it better

than she does:

It is too soon to diagnose with certainty which, if any, of the extant hybrid formswill emerge as successful and whick will fall by the wayside. One pattern thatdoes emerge, however, is striking. The two invaluable contributions a hybridform must make – expanding financing options and providing enforceablecommitments to a blended mission – appear to trade off against each other. ... It ispossible that the impasse cannot be breached, and instead the creators of anyhybrid model will simply have to choose a point at which they are willing to tradeaccess to capital for enforcement of blended mission.

43

References

Bagnoli, Mark, and Susan G. Watts. 2003. “Selling to Socially Responsible Consumers:Competition and the Private Provision of Public Goods.” Journal of Economics andManagement Strategy 12 (3): 419–45.

Baron, David P. 2007. “Corporate Social Responsibility and Social Entrepreneurship.” Journalof Economics and Management Strategy 16 (3): 683–717.

Baron, David P. 2009. “A Positive Theory of Moral Management, Social Pressure, and CorporateSocial Performance.” Journal of Economics and Management Strategy 18 (1): 7–43.

Ben-Ner, Avner, and Derek Jones (1995). Employee Participation, Ownership, and Productivity:A Theoretical Framework. Industrial Relations 34:532–554.

Besley, Timothy, and Maitreesh Ghatak. 2007. “Retailing Public Goods: The Economics ofCorporate Social Responsibility.” Journal of Public Economics 91 (9): 1645–63.

Bilodeau, Marc and Slivinski, Al (1998) Rational Nonprofit Entrepreneurship, Journal ofEconomics and Management Strategy, 7, 551-571.

Brakman Reiser, Dana (2009). For-Profit Philanthropy. Fordham Law Review 77(5): 2437-2473.

Brakman Reiser, Dana (2010). Governing and Financing Blended Enterprise. Chicago-KentLaw Review 85(2): 619-655.

Brekke, Kjell Arne, and Karine Nyborg. 2004. “Moral Hazard and Moral Motivation: CorporateSocial Responsibility as Labor Market Screening.” University of Oslo Department ofEconomics Memorandum 25/2004.

Burlingame, Dwight F. (2004). Andrew Carnegie. In Dwight F. Burlingame, ed., Philanthropyin America. Santa Barbara, CA: ABC-CLIO, Inc.: 59-61.

Callanan, Laura and Jonathan Law. (2012). Will social impact bonds work in the United States? McKinsey and Company. Available online athttp://payforsuccess.org/sites/default/files/will_social-impact-bonds_work_in_the_us.pdf

Calveras, Aleix, Juan-Jose Ganuza, and Gerard Llobet. 2007. “Regulation, Corporate SocialResponsibility and Activism.” Journal of Economics and Management Strategy 16 (3):719–40.

44

Caviccio, Carolyn and Judit Torok (2009). The 2009 Corporate Contributions Report. NewYork: The Conference Board, Report Number R-1458-09-RR.

Clark, William H., Jr., and Larry Vranka,(principal authors). (2012). White Paper. The Needand Rationale for the Benefit Corporation: Why it is the Legal Form that Best Addressesthe Needs of Social Entrepreneurs, Investors, and, Ultimately, the Public. Available athttp://www.benefitcorp.net/storage/The_Need_and_Rationale_for_Benefit_Corporations_-_April_2012.pdf

Clotfelter, Charles T. (1985). Federal Tax Policy and Charitable Giving. Chicago: Universityof Chicago Press.

Doeringer, Matthew F. (2009-10). Note, Fostering Social Enterprise: A Historical andInternational Analysis. Duke Journal of Comparative and International Law 20: 291-329.

Friedman, Milton (1970). The Social Responsibility of Business Is to Increase Its Profits. TheNew York Times Magazine, (13 September), 32–33, 122, 126.

Galaskiewicz, Joseph and Michelle Sinclair Colman (2006). Collaboration betweenCorporations and Nonprofit Organizations. In Walter W. Powell and Richard Steinberg,eds., The Nonprofit Sector: A Research Handbook, Second Edition. New Haven: YaleUniversity Press, pp. 180-198.

Giving USA Foundation (2010). Giving USA. http://www.givingusareports.org/free.php,Accessed Oct. 24, 2010.

Google.org (2012) http://www.google.org/about.html. Various dates, last accessed June 30,2012.

Graff Zivin, Joshua, and Arthur Small. 2005. “A Modigliani–Miller Theory of AltruisticCorporate Social Responsibility.” B. E. Journal of Economic Analysis and Policy: Topicsin Economic Analysis and Policy 5 (1).

Hansmann, Henry (1980). The Role of Nonprofit Enterprise. Yale Law Journal 89:835–901.

Hansmann, Henry (1981) The Rationale for Exempting Nonprofit Corporations from theCorporate Income Tax. Yale Law Journal 91:54-100.

Hansmann, Henry, Daniel Kessler, and Mark McClellan (2003). Ownership Form And TrappedCapital In the Hospital Industry. In Edward Glaeser, ed., The Governance ofNot-For-Profit Organizations. Chicago: University of Chicago Press.

45

Himmelstein, Jerome L. (1997). Looking Good and Doing Good: Corporate Philanthropy andCorporate Power. Bloomington, IN: Indiana University Press.

IRS. (2012). Examples of Program-Related Investments. Document IRS-2012-0015-0001,Posted online and in the Federal Register on April 19, 2012. Available atwww.regulations.gov.

Irvin, R. A., & Carr, P. (2005). The Role of Philanthropy in Local Government Finance. PublicBudgeting & Finance, 25(3), 33-47. doi:10.1111/j.1540-5850.2005.00366.x

Katz, Robert A. and Antony Page. (2009). The Role of Social Enterprise. Vermont Law Review35: 59-103

Kelley, Thomas A. (2009). Law and Choice of Entity on the Social Enterprise. Tulsa LawReview 84. Available at: http://works.bepress.com/thomas_kelley/2

Kitzmueller, Markus and Jay Shimshack (2012) Economic Perspectives on Corporate SocialResponsibility. Journal of Economic Literature L#1: 51-84

Kotchen, Matthew J. 2006. “Green Markets and Private Provision of Public Goods.” Journal ofPolitical Economy 114 (4): 816–34.

Krashinsky, Michael. (1986). Transaction Costs and a Theory of the Nonprofit Organization. InSusan Rose-Ackerman, ed., The Economics of Nonprofit Institutions. New Haven: YaleUniversity Press, pp. 114-132.

Li, Sherry Xin , Catherine C. Eckel, Philip J. Grossman, Tara Larson Brown, Giving togovernment: Voluntary taxation in the lab, Journal of Public Economics, Volume 95,Issues 9–10, October 2011, Pages 1190-1201.

Lyon, Thomas P., and John W. Maxwell. (Forthcoming). Greenwash: Corporate EnvironmentalDisclosure under Threat of Audit. Journal of Economics and Management Strategy.

Mac Cormac, Susan, with assistance from Jonathan Glass and Julie Cooke. (2007). TheEmergence of New Corporate Forms: The need for alternative corporate designsintegrating financial and social missions. In Allen White and Marjorie Kelley, eds. PaperSeries on Corporate Design. Available online at www.corporation2020.org/pdfs/SummitPaperSeries.pdf

Mackey, J., 2005, “Putting Customers Ahead of Investors,” Reason,www.reason.com/0510/fe.nf.rethinking.shtml

Maxwell, John W., Thomas P. Lyon, and Steven C. Hackett. 2000. “Self-Regulation and Social

46

Welfare: The Political Economy of Corporate Environmentalism.” Journal of Law andEconomics 43 (2): 583–617.

Navarro, Peter (1988). Why Do Corporations Give to Charities? Journal of Business 61(1): 65-93.

Page, Antony and Robert A. Katz. (2010). Freezing out Ben & Jerry: Corporate Law and theSale of a Social Enterprise Icon. Vermont Law Review, 35: 211-250.

Posner, Eric A. and Anup Malani. (2007). The Case for For-Profit Charities. Virginia LawReview 93: 2017-2067.

Posnett, John W. and Todd Sandler. 1986. Joint Supply and the Finance of Charitable Activity. Public Finance Quarterly 14: 209-222.

Porter, M. E. and M. R. Kramer (2002). The Competitive Advantage of Corporate Philanthropy.Harvard Business Review 80(12): 56–68.

Preston, Anne E. (1989) The Nonprofit Worker in a For-Profit World, Journal of LaborEconomics, 7, 438-463

Python, Monty. (1975). Monty Python and the Holy Grail (film). Relevant segment availableonline at www.youtube.com/watch?v=dGFXGwHsD_A

Regulator of Community Interest Companies. (2009). Frequently Asked Question 13. Availableat www.cicregulator.gov.uk/CICleaflets/FAQ.

Rodgers, T.J. (2005). Put Profits First. Reason, www.reason.com/0510/fe.nf.rethinking.shtml

Schlesinger, Mark and Bradford H. Gray. (2006). Nonprofit Organizations and Health Care:Some Paradoxes of Persistent Scrutiny. In Walter W. Powell and Richard Steinberg, eds.,The Nonprofit Sector: A Research Handbook, Second Edition. Yale University Press:378-414.

Slivinski, Al. (2002). Team Incentives and Organizational Form. Journal of Public EconomicTheory 4:185–206.

Social Finance, Inc. (2012). A New Tool for Scaling Impact: How Social Impact Bonds CanMobilize Private Capital To Advance Social Good. Available on-line athttp://www.rockefellerfoundation.org/uploads/files/655fab01-83b9-49eb-b856-a1f61bc9e6ca-small.pdf

Social Investment Forum Foundation. (2010). Report on Socially Responsible Investing Trends

47

in the United States 2010, Executive Summary. Available on-line athttp://ussif.org/resources/pubs/trends/documents/2010TrendsES.pdf

Starke, Christoph. 2012. Serving the Many or Serving the Most Needy? (Either a workingpaper, or possibly will be accepted for publication in Economics of Governance (checkbefore copyediting).

Steinberg, Richard. (1987). Nonprofits and the Market. In Walter W. Powell, ed., TheNonprofit Sector: A Research Handbook. New Haven: Yale University Press: 118-140.

Steinberg, Richard. (2006) Economic Theories of Nonprofit Organizations. In Walter W.Powell and Richard Steinberg, eds., The Nonprofit Sector: A Research Handbook, SecondEdition. Yale University Press: 117-139.

Steinberg, Richard and Burton A. Weisbrod (1998). Pricing and Rationing by NonprofitOrganizations with Distributional Objectives. In Burton A. Weisbrod, ed., To Profit orNot to Profit: The Commercial Transformation of the Nonprofit Sector. New York:Cambridge University Press, pp. 65-82.

Steinberg, Richard and Burton A. Weisbrod (2005). Nonprofits with Distributional Objectives:Price Discrimination and Corner Solutions. Journal of Public Economics, 89: 2205-2230.

Strom, Stephanie and Miguel Helft. (2011). Google Finds it Hard to Reinvent Philanthropy. TheNew York Times (January 29).

The Social Investment Consultancy (2010) Press release, March 23, 2010. http://www.tsiconsultancy.com/corporate-giving-to-be-cut-by-a-third/, last accessed Nov.3, 2010.

Webb, Natalie J. and Amy Farmer (1996). Corporate Goodwill: A Game Theoretic Approach tothe Effect of Corporate Charitable Expenditures on Firm Behavior. Annals of Public andCooperative Economics 67:29-50.

Wedig, Gerard (1994) Risk, Leverage, Donations and Dividends-in-Kind: A Theory ofNonprofit Financial Behavior, International Review of Economics and Finance, 3, 257-278.

Weisbrod, Burton (1975) Toward a Theory of the Voluntary Nonprofit Sector in a Three-SectorEconomy. In Edmund S. Phelps, ed., Altruism, Morality, and Economic Theory. NewYork: Russell Sage Foundation.

Wikipedia (2012). http://en.wikipedia.org/wiki/Takeovers. Last accessed June 28, 2012.

48

Yunus, Muhammad (2010). Excerpt from Building Social Business: The New Kind of CapitalismThat Serves Humanity's Most Pressing Needs. as cited athttp://www.muhammadyunus.org/In-the-Media/social-entrepreneurs-must-drop-the-balancing-act/, June 9, 2010. Accessed Nov. 5, 2010.

49